An Introduction

to
Computer Science
Robert Sedgewick
Kevin Wayne
Princeton University

Copyright 2003 by Robert Sedgewick and Kevin Wayne

For information on obtaining permission for use of material from this work, please submit
a request to the authors at rs@cs.princeton.edu and wayne@cs.princeton.edu

Preface

Computer science as an academic discipline has evolved to embrace a set of

intellectual challenges on a par with other sciences. This fact, combined with
the undeniable impact of computer science on the modern world, demands
an introductory college textbook comparable with commonly-used textbooks in physics, chemistry, or biology. Accordingly, this book is intended to
meet the need for an introductory college text in computer science. The distinctive feature of the book is that it has broader coverage of the field than is
found in many texts that are currently in use.
All college students can benefit from exposure to computer science
early in the curriculum. Therefore, increasing numbers of colleges and universities are requiring that students (particularly those in the sciences, engineering, mathematics, and even the social sciences) take one or two semester
courses in computer science in the first year. In more mature disciplines, it is
commonly accepted that, at this critical point in the curriculum, it is important to challenge students with fundamental intellectual issues while surveying the field. This book aims to support an introductory course that does so
for computer science.
We have three primary goals. First, we want to demystify computer
systems by unpeeling levels of abstraction down to the simplest physical component, leaving no black boxes. Second, we want to empower students by giving them the experience and insight necessary to exploit available technology
whenever appropriate. Third, we want to build awareness of the substantial
intellectual underpinnings of the field and its broad reach into other sciences.

ii

The book is a self-contained treatment intended for people with no

previous experience in computer science. While its primary purpose is to
serve as a textbook for first-year college students, it may also serve as a broad
introduction to the field that may be of interest to anyone who has not been
exposed to its fundamentals.

Coverage
The book is organized around four areas of computer science: programming,
architecture, theory, and systems. We introduce fundamental concepts in each
area and pay special attention to relationships among them. A proper introduction to the field must do justice to each of these four areas. We also cover
applications in scientific computing throughout the book, to reinforce and
supplement the curriculum that is typically taught to students in mathematics,
science, and engineering in high school and their first year in college.
Programming is essential to being able to understand and appreciate
computer science. We cover basic programming in Java, abstract data types
and Java classes, elementary data structures, and the design and analysis of
algorithms. To teach these concepts, we use sample programs that solve
interesting problems supported with exercises ranging in difficulty from selfstudy drills to challenging problems that call for creative solutions. Whenever
possible, we draw examples from concepts to be covered elsewhere in the
book or from scientific or commercial applications. Students who learn the
programs in this book will have the solid foundation necessary to prepare
them to be able to effectively use computers as they pursue any academic discipline.
Architecture refers both to the art of designing computers and to the
process of building them. Our coverage of this topic is centered around an
imaginary machine that is similar to real computers. We specify the machine
in full detail, consider machine-language programs for familiar tasks, and
present a Java simulator for the machine. We continue with a treatment of
circuits and logical design, culminating in a description of how such a
machine might be built from the ground up. Our intent is to demystify this
apparently formidable challenge while also providing a context for under-

An Introduction to Computer Science

iii

standing how Java programming and other familiar high-level tasks relate to
actual machines.
Theory helps us to understand the limits on what we can accomplish
with computers. We present Turings classical results that show how simple
abstract machines can help us to pose fundamental questions about the
nature of computation. Some of these are among the most important scientific questions of our time. We also consider practical applications such as
how to estimate the running times of our programs and how to design efficient algorithms.
Systems enable us to work with computers at a high level of abstraction. We describe the basic components of computer systems that support
programming; operating systems for interacting with our programs and our
data; networks that allow interaction among computers; and applications
systems that provide specialized support for particular tasks.
Our coverage of the four areas is intertwined and also threaded with
descriptions and examples of various classical algorithms, programming languages, scientific computing, and commercial applications. Where relevant,
we also provide proper historical perspective.
Generally speaking, we introduce material that is covered, at most
colleges and universities, in several later computer science courses. For computer science majors, this breadth of coverage provides a foundation to be
reinforced in later courses. For students in other fields who have a chance to
take only one or two courses in computer science, this introduction to programming, architecture, theory and systems in the context of scientific applications gives the background that they need to effectively address the impact
of computer science on their chosen fields.

Use in the Curriculum

This book is intended for a college course aimed at teaching novices
to program while at the same time introducing them to the field of computer
science. As such, its content fulfills the requirements of a first course in computer science in a reasonable way. But it does offer an alternative to many
courses that are traditionally taught. Rather than delving deeply into the
details of a particular programming language, we put programming in con-

iv

text. Rather than giving scant coverage to programming while surveying

computer science, we are able to address fundamental issues that cannot be
understood without programming experience.
One option is to use this book is as the basis for a full-year college
course, perhaps supplemented with a standard text on computer programming. Depending on the needs of the students and the experience of the
instructor, diversions of varying length on basic programming skills are
appropriate. For logical consistency, we have put most of the material on
programming at the beginning of the book; in practice, it is best to spread the
coverage of programming throughout the course, using weekly programming
assignments tied to the material in the text.
Another option is to use the book as the basis for a fast-paced onesemester course. This choice is appropriate in situations where students in
the sciences and engineering may have only one semester to devote to computer science. The course is best positioned early in the curriculum, because
students who take it are sure to be able to use computers much more effectively when necessary in courses in their specialty. If the course is later in the
curriculum, students may have a more mature point of view that better
enables understanding of the advanced topics (but not much patience with
learning the basics of programming).
Simply put, this book provides the following alternative to traditional introductory courses: a way to introduce students to computer science
while at the same time teaching them to program, in a single course.

Prerequisites
Our aim is for the book to be suitable for typical first-year science
and engineering students. Accordingly, we do not expect preparation significantly different from other entry-level science and mathematics courses.
Learning to program is one of our primary goals, so we assume no
prior programming experience. Indeed, one of the most important features
of our approach is that we integrate teaching programming with teaching the
rest of computer science. But what should be done with students who have
already taken a programming course, in high school or in college? Actually
the material in this book is ideal for such students. They can review the mate-

An Introduction to Computer Science

rial on programming (which is likely to be in a different context from what

they learned) and focus on the intellectual issues at the heart of the book
(which are not likely to have been covered at all in their earlier course).
There is no substitute for experience when learning to program. Anyone writing a program to solve a new problem faces a challenging intellectual
task, just as does anyone writing an essay on a new topic. Students who have
written numerous essays in high school still benefit from introductory writing courses in college; just as students who have written numerous programs
in high school can benefit from taking an introductory programming course.
This analogy breaks down slightly because no one comes to college never having written an essay, but many students come to college never having written
a program. But our experience has been that we can get students to the point
where they can confidently write a program relatively quickly and that virtually all students in science and engineering can benefit from taking this as a
first course (perhaps classified according to their programming experience).
Mathematical maturity is just as important. While we do not dwell on
mathematical material, we do refer to the mathematics curriculum that students have taken in high school, including algebra, geometry, and trigonometry. Most students in our target audience (those intending to major in the
sciences and engineering) automatically meet these requirements. Discrete
mathematics plays a critical role in computer science but is not always fully
covered in mathematics curricula, so we cover topics such as Boolean logic,
mathematical induction, basic probability, and discrete sums with the point
of view that most students have some familiarity with them but that the
major concepts need to be reinforced (and we include an appendix with some
basic information on discrete mathematics).
We also describe numerous scientific applications, integrated thoughout the text. We occasionally draw on examples from probability, statistics,
and calculus and Chapter 9, on scientific computing, covers several advanced
topics. This material can be skipped by students who do not have requisite
preparation (though they may wish to refer to it when they do take the more
advanced courses). Otherwise, our examples do not assume any preparation
beyond that provided by typical high-school courses in mathematics, physics,
biology, or chemistry.

vi

Goals
What can instructors of upper-level courses in science and engineering expect
of students who have successfully completed a course based on this book?
Anyone who has taught an introductory computer science course
knows that expectations are typically high: each instructor expects all students to be familiar with the computing environment and approach that he
or she wants to use. A physics professor might expect some students to learn
to program over the weekend to run a simulation; an engineering professor
might expect other students to be using a particular package to numerically
solve differential equations; or a computer science professor might expect
knowledge of the details of a particular programming environment. Is it realistic to expect to be able to meet such diverse expectations? Should there be a
different introductory course for each set of students? Colleges and universities have been wrestling with such questions since computers came into widespread use in the latter part of the 20th century.
Our primary goal is to provide an answer to such questions by developing a common introductory course in computer science for all students in
science and engineering (including prospective computer science majors)
that is analogous to commonly-accepted introductory courses in mathematics, physics, biology, and chemistry. The course may also be suitable for students in the humanities who wish a full introduction to the field.
Students who master the material in this book gain the confidence
and knowledge that they need to be able to learn to exploit computers wherever they might appear later in their careers, whether it be using an integrated
mathematical software package to attack advanced mathematical problems,
using Java to develop innovative applications, writing code to control sophisticated devices, using simulation to study complex problems, or developing
new computational paradigms. People continually need to be able to develop
new skills in particular contexts because computers continue to evolve.
Instructors teaching students who have studied from this book can expect
that they have the background and experience necessary to make it possible
for them to acquire such skills, to effectively exploit computers in diverse
applications, and to appreciate their limitations.

An Introduction to Computer Science

vii

Booksite
An extensive amount of information that supplements this text may be found
on the world-wide web at
http://www.cs.princeton.edu/IntroCS

For economy, we refer to this web site as the booksite throughout. It contains
material oriented towards instructors, students, and casual readers of the
book. We briefly describe this material here, though, as all web users know, it
is best surveyed by browsing. With a few exceptions to support testing, the
material is all publicly available.
One of the most important implications of the booksite is that it
empowers instructors and students to use their own computers to teach and
learn the material in this course. Anyone with a computer and a browser can
begin learning computer science and learning to program, by following a few
instructions on the booksite. The process is no more difficult than downloading a media player or a new computer game.
For instructors, the booksite contains information about teaching the
course. This information is primarily organized around a teaching style that
we have developed over the past decade, where we offer two lectures per week
to a large audience, supplemented by two class sessions per week where students meet in small groups with instructors or teaching assistants. The booksite has presentation slides for the lectures, which set the tone for the course.
We assign weekly problem sets based on exercises from the book and
programming assignments, also based on exercises from the book, but with
much more detail. Each programming assignment is intended to teach a relevant concept in the context of an interesting application while presenting an
ambitious and interesting challenge to each student. The progression of
assignments embody our approach to teaching programming. The booksite
fully specifies all the assignments and provides detailed, structured information to support teaching students to complete them in the alloted time,
including code for strawman solutions, descriptions of suggested approaches,
and outlines for what should be taught in class sessions.
The booksite also includes webware for use in managing student submissions and grading assignments.

viii

For students, the booksite contains quick access to much of the material in the book, plus extra material to encourage self-learning. Solutions are
provided for many of the books exercises, including complete program code
and text data. There is a wealth of information associated with programming
assignments, including suggested approaches, checklists, FAQs, and test data.
For casual readers (including instructors and students!) the booksite
is a resource for accessing all manner of extra information associated with the
books content. All of the booksite content provides web links and other
routes to pursue to find more information about the topic under consideration. There is far more information accessible than any individual could
fully digest, but our goal is to provide enough to whet any readers appetite
for more information about the books content.
As with any web site, our Introduction to Computer Science booksite is
continually evolving, but it is an essential resource for everyone who owns
this book. In particular, the supplemental materials supporting teaching and
learning within a first-year college course are critical to our goal of making
computer science an integral component of the education of all scientists and
engineers.

Acknowledgements

An Introduction to Computer Science

ix

Contents
1

Overview . . . . . . . . . . . . . . . . . . .1
1.1
1.2
1.3
1.4

Elements of Programming . . . . . . . . . 21
2.1
2.1
2.2
2.3
2.4
2.5
2.6

A Simple Machine 1
Applications 1
Analysis 1
Context 1

Your First Program 22

Primitive Types of Data 21
Conditionals and Loops 43
Arrays 79
Functions (static methods) 80
Recursion 81
Input and Output 82

Object-Oriented Programming . . . . . . 137

3.1
3.2
3.3
3.4

Data Types and Java Classes 137

Modular Programming 137
Encapsulation and ADTs 137
Inheritance 137

Fundamental ADTs . . . . . . . . . . . . 185

4.1
4.2
4.3
4.4
4.5

A Computing Machine . . . . . . . . . . 239

5.1
5.2
5.3
5.4
5.5

Data Representations 239

TOY machine 239
Instruction Set 239
Machine-Language Programming
TOY Simulator 239

239

Building a Computer . . . . . . . . . . . 283

6.1
6.2
6.3
6.4
6.5

Linked Structures 185

Stacks and Queues 185
Priority Queues 185
Symbol Tables 185
Graphs 185

Boolean Logic and Gates 283

Combinational Circuits 283
Sequential Circuits 283
Components 283
TOY Machine Architecture 283

Theory of Computation . . . . . . . . . . 341

7.1
7.2
7.3
7.4

Languages and Finite-State Automata 341

Turing Machines 341
General-Purpose Computers 341
Computability 341

An Introduction to Computer Science

xi

7.5 Chomsky Hierarchy 341

7.6 Proving Properties of Programs 341

Systems . . . . . . . . . . . . . . . . . . 387
8.1
8.2
8.3
8.4
8.5

Scientific Computation . . . . . . . . . . 441

9.1
9.2
9.3
9.4
9.5
9.6

10

Library Programming 387

Compilers, Interpreters, and Emulators 387
Operating Systems 387
Networks 387
Applications Systems 387

Precision and Accuracy 441

Symbolic Methods 441
Linear Algebra 441
Solution of Differential Equations
Data Analysis 441
Simulation 441

441

Analysis of Algorithms . . . . . . . . . 487

10.1
10.2
10.3
10.4
10.5
10.6

Predicting Performance 487

Guaranteeing Performance 487
Reduction 487
Computational Complexity 487
Intractability 487
Case Studies 487

xii

An Introduction to Computer Science

Overview

1.1

A Simple Machine

1.2

Applications

1.3

Analysis

1.4

Context

Overview

An Introduction to Computer Science

Elements of Programming

Our goal in this chapter is to convince you that writing a program is easier
than writing a piece of text such as a paragraph or an essay. Writing prose is
difficult: we spend many years in school to learn how to do it. By contrast, just
a few building blocks suffice to take us into a world where we can harness the
computer to help us solve all sorts of fascinating problems that would be otherwise unapproachable. In this chapter, we take you through these building
blocks, get you started on programming in Java, and study a variety of interesting programs. You will have an additional avenue to be able to express yourself
(writing programs) within just a few weeks. Like the ability to write prose, the
ability to program is a lifetime skill that you can continually refine and is certain to serve you well into the future.
You will learn the Java programming language, but that will be much
easier to do than learning a foreign language that is unfamiliar to you.
Indeed, programming languages are characterized by no more than a few
dozen vocabulary words and rules of grammar. Most of the material that we
cover in this chapter could apply to the C or C++ languages, or any of several
other modern programming languages, but we describe everything specifically in Java so that you can get started creating and running programs right
away. On the one hand, to the extent possible, we will focus on learning to
program, as opposed to learning details about Java. On the other hand, part
of the challenge of learning to program is knowing which details are relevant
in a given situation. Java is widely available, but also learning to program in
Java will make it easy for you learn to program in another language.

21

Elements of Programming

22

2.1

Your First Program

In this section, our plan is to lead you into world of Java programming by taking you through the basic steps required to get a simple program running. The
Java system is a collection of applications not unlike any of the other applications that you are accustomed to using (such as your word processor, e-mail
program, or internet browser). As with any application, you need to be sure
that Java is properly installed on your computer. It comes preloaded on may
computers and is easy to download. You also need an editor and a terminal
application (see Appendix X).
Programming in Java. To introduce you to developing computer programs written in Java, we break the process down in to three steps. To program in Java you need to:
1.

create the program by typing it into a file named, say, MyProgram.java

2.

compile it by typing javac MyProgram.java in a terminal window

3.

run (or execute) it by typing java MyProgram

In the first step, you start with a blank page and end with a sequence of typed
characters on the page, just as when you write an e-mail or a paper. In the second step, you use a system application called a compiler that translates your
program into a form more suitable for the computer (and puts the result in a
file named MyProgram.class). In the third step, you transfer control of the
computer from the system to your program (which returns control back to
the system when finished).
Creating a program. A program is nothing more than a sequence of characters, like a sentence, a paragraph, or a poem. To create one, therefore, we
need only define that sequence of characters, in the same way as we do for email or any other computer application. Programmers usually use a simple
editor for this task with a fixed-width font to make it easier to line things up
vertically in the code.
Compiling a program. At first, it might seem to you as though the Java
programming language is designed to be best understood by the computer.
Actually, to the contrary, the language is designed to be best understood by

AN INTRODUCTION TO COMPUTER SCIENCE

Your First Program

the programmer (thats you). The computers language is more primitive

than Java, as we shall see in Chapter 5. A compiler is an application that translates program from the Java language to a language more suitable for executing on the computer. It takes a file with a . java extension as input (your
program) and produces a file with the same name but with a . class extension (the computer-language version). To use the compiler to compile a program, type the javac command followed by the program name in a terminal
window. Most systems have other ways to interact with the compiler; we
choose this one here because its use is simple and compact to describe.
Executing a program. Once it has been compiled, we can run the program. This is the exciting part, where your program actually takes control of
your computer (within the constraints of what the Java system allows). It is
perhaps more acurate to say that the computer follows your instructions.
Program 2.1.1 is an example of a complete Java program. Its name is
HelloWorld, so that its code must reside in a file named HelloWorld.java
(by convention in Java). The programs sole action is to print a message back
to the terminal window. For continuity, we will use some standard Java terms
to describe the program without defining them until Section X, but you do
not now need to know details of these definitions: Program 2.1.1 consists of a
single class named HelloWorld that has a single method named main() that
uses a method named println() from Javas System.out library to do the
job. for the time being, you can think of class as meaning collection of
programs and method as meaning program. When referring to methods, we use () after the name to distinguish method names from other kinds
of names.
Since the 1970s, it has been a tradition that everyone learning to program should start with HelloWorld, so you should type it into a file, compile
it, and run it. By doing so, you will not just be following in the footsteps of
countless other people who have learned how to program, but you also will
be checking that you have a usable editor and terminal emulator (see Appendix X) and that your Java system is properly installed.
A method comprises a signature, which has its name and other information, and a block, which is sequence of statements enclosed in braces and
each followed by a semicolon. The statements in a methods block are exe-

23

Elements of Programming

24

Program 2.1.1 Hello, World

public class HelloWorld
{
public static void main(String[] args)
{
System.out.println(Hello, World);
}
}

This code is a complete Java program

that accomplishes a simple task
(printing a message on the terminal
window). It is traditionally a beginning programmers first program .

% javac HelloWorld
% java HelloWorld
Hello, World

cuted, one by one, when the method is invoked. One type of statement is a
method name with, in parentheses, zero or more arguments. When we write
a such a statement, we are simply saying that we want to run that method
(and to provide it with some information, in the arguments). This process is
known as invoking or calling the method. In HelloWorld, the main method
consists of a single statement that calls the Java library method System.out.println(). The name of the method is println(); the name of
the library is System.out. You will be writing programs that use calls to many
different Java library methods, and you refer to each of them in just this way.
In Program 2.1.1, System.out.println() takes the message Hello, World
as its argument and prints it to the terminal. We do not need to know the
details of how System.out.println() accomplishes this taskwe can simply rely on it to do so. The method main() itself takes an argument, which we
will discuss soon as the focus of Program 2.1.2.
At first, accomplishing the task of printing something out in a terminal window might not seem very interesting; upon reflection, you will see
that one of the most basic capabilities that we need from a program is for it to
be able to tell us what it is doing.
For the time being, all our program code will be just like
Program 2.1.1, except with a different sequence of statements in main(). We

AN INTRODUCTION TO COMPUTER SCIENCE